NEW MODELS OF OCULAR INFLAMMATION AND NEW WAYS TO STUDY THEM: (1) Birdshot chorioretinopathy (BC) is strongly associated with HLA-A29. To study the association between HLA-A29 and uveitis, we applied for and received a Bench to Bedside award for FYs 2011-13. As part of the project we developed HLA-A29 transgenic mice using a construct obtained from collaborators. Although the mice developed ocular pathology, they did not express HLA-A29 protein and the pathology turned out to be due to the hitherto unknown presence of the rd8 mutation of the Crb1 gene in the C57BL/6N strain. After breeding out the mutation, pathology is not apparent. We are currently preparing a new HLA-A29 construct to rederive this strain. (2) We continue to collaborate with Drs. Warren Strober (NIAID) and Ivan Fuss on the study of ocular inflammation in mice with an inflammasome mutation. These mice are knock-in for a mutated NLRP3 gene associated with Muckle-Wells syndrome, which among its pathologies is also associated with inflammation at the ocular surface. We developed a new model of ocular surface inflammation by instilling drops containing a mixture of microbial stimulants and found effects on mucosal immune responses at the ocular surface, in the mutant, but not in wild type mice. This may provide new insights into ocular involvement in Muckle-Wells syndrome. (3) We have derived three lines of T cell receptor (TCR) Tg mice that express different copy numbers of a TCR specific to the major epitope of IRBP, named (from high expression to low) R161H, R161M and R161L. R161H and to a lesser extent R161M mice develop spontaneous uveitis by 2 months of age. R161L mice develop minimal or no spontaneous disease. These mice thus represent a new model system that permits to study mechanisms of spontaneous uveitis and serve as a source of retinal antigen-specific T cells, allowing for the first time to study their development and function. Using these mice, insights are being gained into basic mechanisms of autoimmune uveitis. (Horai et al., J. Autoimmun. 2013) (4) We performed comparative analysis of three models of uveitis: induced and two spontaneous. We merged several approaches to analyze disease development and progression, including fundus examination and photography using Micron III, optical coherence tomography (OCT), electroretinography (ERG) and finally histopathology. New information about pathogenesis of EAU in these models was obtained, and a previously unappreciated chronic variant of EAU in B10.RIII mice was identified (Chen et al., PloS One 2013 and Chen et al., PloS One, in press). FUNDAMENTAL MECHANISMS IN TOLERANCE, IMMUNITY AND AUTOIMMUNITY TO RETINAL ANTIGENS (1) Spontaneous uveitis was ameliorated in R161H mice treated with a broad-spectrum antibiotic mixture to eliminate endogenous bacterial flora or reared under germ free conditions. This suggests that commensals contribute to development of spontaneous ocular autoimmune disease. IL-17-producing TCR Tg R161H T cells were enriched in the lamina propria of the gut of unmanipulated mice and data indicate presence of a TCR-driven response by R161H cells in the gut. We hypothesize that R161H cells are activated by bacterial flora and migrate to the eye to elicit uveitis (Horai et al, manuscript in preparation). We are currently attempting to dissect which component(s) of the flora is(are) responsible. (2) crossing R161H mice to IFN-g deficient or IL-17 deficient mice unexpectedly revealed a major role for IFN-g in the spontaneous disease. IFN-g-/- R161H mice, but not IL-17-/- R161H mice, had severely reduced spontaneous uveitis scores. These findings raise the possibility that IL-17 produced by R161H cells in the gut, as described above, is a marker of the pathogenic cells but not necessarily the pathogenic cytokine itself. (3) We previously demonstrated the importance of Vitamin A (VitA) and its metabolite, retinoic acid, in ocular immune privilege. Using mice made VitA deficient (VAD) and employing the induced and the spontaneous models of EAU, we are studying the role of VitA in regulation of autoimmunity to retina. We found that although nave T cells in VAD mice display a defect in activation, T cell effector function that was acquired before onset of VAD is fully maintained in the VAD host and can actually drive enhanced disease, possibly due to reduced ability to generate Treg cells under conditions of VitA deficiency. These findings may have clinical implications in geographical regions where dietary VitA is limiting. (Zhou et al, in preparation) (4) Autoimmunity to retina could be either Th17 or Th1 driven. Because immune responses have inherent plasticity, targeting only one response could drive the response towards the other, while continuing to fuel pathology. IL27p28 (binds to gp130, a component of both Th17-promoting IL-6, and Th1-promoting IL-27 receptors) was able to inhibit both responses concurrently, by affecting effector priming and lineage stability. Furthermore, similar effects of IL-27p28 in vitro are seen on human T cells. This could point to IL-27 and/or its subunits as a therapeutic mechanism in immunologically complex diseases such as uveitis. (Chong et al, J Autoimm., in press). (5) IL-22 has variably been reported to have both pro-inflammatory and protective effects, depending on the tissue and the model. We used IL-22 and IL-22-receptor deficient mice and anti-IL-22 antibodies to examine effects of IL-22 modulation on EAU. Our data suggest that IL-22 has a local anti-inflammatory and tissue-protective role in the eye (Mattapallil et al., in preparation). (6) We examined whether T regulatory (Treg) cells found in uveitic eyes are (i) IRBP specific, (ii) functionally suppressive, and (iii) may play a role in natural resolution of disease. Using IRBP-MHC dimers as well as FoxP3 reporter and FoxP3 deleter mice, we found that the T cell infiltrate in uveitic eyes of mice that have a polyclonal T cell repertoire is highly enriched in IRBP-specific Treg and Teff cells. Unlike what has been reported for Treg in other inflammatory sites, Treg in uveitic eyes appear unimpaired functionally. Finally, FoxP3+ Treg appear to play a role in the natural resolution of EAU and in the maintenance of remission, at least in part due to local effects within the eye (Silver et al, in preparation). (7) In chronically inflamed eyes of R161H mice we identified structures resembling tertiary lymphoid tissue. Using confocal microscopy and local delivery of antibodies, these structures are being studied for their cellular composition, evidence of local T cell priming, and whether they required for persistence of disease. (J. Kielczewski, in preparation) EFFECTS OF INNATE IMMUNE RESPONSES ON AUTOIMMUNITY: The innate immune response directly affects immunopathogenic processes and also impacts on adaptive immunity. (1) We previously identified a population of NKT cells that express the IL-23R constitutively and produce IL-17 independently of IL-6 and IL-21 (NKT17). Recent data indicate that IL-17 production in these cells may involve a unique signaling pathway that bypasses STAT3 under some conditions. We currently are gathering in vivo data to examine whether there is a role for this pathway in vivo. (A. Hansen, T. St.Leger, in preparation). (2) The cytokine IFN-g can be either protective or proinflammatory in autoimmunity, which is an unsolved paradox. Our recent data suggest that there is a positive feedback loop between NK cell derived IFN-g acting on DC to produce IL-27, which in turn augments the IFN-g; response. This is in line with earlier data published by our lab (Grajewski, Hansen et al., J Immunol 2008) that innate IFN-g; production dampens subsequent adaptive IFN-g and IL-17 responses and protects from EAU (Chong et al, in preparation).